限行条件下配送车辆配置的二阶段随机规划

Stochastic programming for delivery fleet planning under vehicle operational restrictions

针对车辆限行下配送车辆数量和配送时间均具有不确定性这一难题,通过建立二阶段随机规划模型,实现物流配送车队中传统能源和新能源车辆比例和数量的合理配置,以保障后续生产配送的协同进行.以最小化日均车辆采购成本和后续生产配送期望成本为目标建立二阶段随机整数规划模型;为求解模型,提出改进的随机分支定界算法,并嵌入局部搜索以提高上界的估计效率;通过数据实验验证了模型和算法的有效性.此外,利用该模型和算法,探讨了不同车辆限行措施的执行类型和执行概率对企业车辆配置决策的影响.本文研究有利于解决车辆限行导致的城市物流配送难题,为保障我国城市物流相关企业日常运营提供决策方法.

In many Chinese metropolitan areas, in addition to normal traffic uncertainties, the operability of each vehicle in the fleet can be uncertain due to randomly occurring severe weather condition(e.g.,smog), which leads to imposition of vehicle operational restriction. Considering these uncertainty sources,we propose a 2-stage stochastic integer programming model for the strategic-level delivery fleet planning problem, to ensure the coordination of daily production and delivery operations. The model aims to minimize the strategic-level procurement spending on delivery fleet and the expected operational cost under logistic uncertainty. In order to solve this problem, we adapt a stochastic branch-and-bound(SBB)method, for which a local search heuristic is applied to improve the efficiency of the upper bounding.We also conduct the numerical study to provide the model validity and algorithm viability. Compared to solving the expected value problem, our 2-stage stochastic integer programming model can modestly reduce the average daily operating cost. Finally, we report sensitivity analysis results on how the fleet planning decisions would be affected by different types of vehicle operational restrictions with varying occurrence frequencies. This study offers a solution for city delivery problems caused by the vehicle operational restrictions, which is helpful for improve the daily operating performance of city logistics industry.